How the calculation works
ROI is the gain relative to what you put in: subtract the investment cost from the expected benefit, divide by the cost, and express the result as a percentage. A positive figure means the benefit exceeds the cost on paper; a negative figure means the investment destroys value even before risk is considered. The arithmetic is the easy part. The two inputs are where every serious argument happens.
A worked example
Suppose a new order-management system costs €200,000 all-in and the business case claims €260,000 of benefit over the period you are measuring. The gain is €260,000 − €200,000 = €60,000. Divide by the €200,000 cost and you get 0.30, so the ROI is 30%. The same arithmetic at €190,000 of benefit gives −5%: a €70,000 swing in a benefit estimate, well within normal forecasting error, moves the case from approval to rejection.
What the percentage hides
An ROI figure looks precise, which is exactly why it deserves suspicion. Before anyone quotes this number in an approval paper, pressure-test what it assumes away.
- Benefits are booked as certain. The formula treats a forecast as a fact, and most forecasts in approval papers were written by the team that wants the approval.
- Costs rarely include internal time: the project team, the operational disruption, the managers pulled into workshops for six months.
- There is no time value of money. €260,000 arriving in year three is treated the same as €260,000 arriving next quarter.
- There is no risk adjustment. A safe 20% and a speculative 40% are not comparable, but ROI makes them look as if they are.
Limitations
This calculator does one division. It is not a substitute for a cash-flow model, a sensitivity analysis or a finance review, and it cannot tell you whether the benefit estimate was built up from evidence or backed into from a target. Its useful function is to make the claimed return explicit enough to argue about.